trafficserver-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rayan Arasteh <ryan.aras...@gmail.com>
Subject Re: WebProxyCacheTuning
Date Sat, 14 Nov 2015 04:49:47 GMT
Helo sir:
thanks for your reply.
where can i find these codes? on squid.blog or error.log?

On Sat, Nov 14, 2015 at 3:40 AM, Bryan Call <bcall@apache.org> wrote:

> Have you looked at your response codes to see what the breakdown is?
>
> -Bryan
>
> On Fri, Nov 13, 2015 at 1:26 AM, Rayan Arasteh <ryan.arasteh@gmail.com>
> wrote:
>
>> I have installed ATS 6.1.0 as tr-full for caching my clients Web Traffic
>> on port 80 this week and every thing works fine. but while I'm looking
>> statistics i found something abnormal about* 35% Error-Total*:
>>
>> Request Result                         Count    Percent       Bytes
>> Percent
>>
>> ------------------------------------------------------------------------------
>> Cache hit                              4,709      0.96%    233.83MB
>> 1.62%
>> Cache hit IMS                          3,199      0.65%      2.55MB
>> 0.02%
>> Cache hit refresh                      1,086      0.22%      3.47MB
>> 0.02%
>> Cache hit other                        7,199      1.47%    120.43MB
>> 0.84%
>> Cache hit total                       16,193      3.30%    360.29MB
>> 2.50%
>>
>> Cache miss                           258,347     52.60%      8.53GB
>> 60.66%
>> Cache miss IMS                        23,309      4.75%     52.38MB
>> 0.36%
>> Cache miss refresh                     1,808      0.37%      5.03MB
>> 0.03%
>> Cache miss other                           0      0.00%      0.00KB
>> 0.00%
>> Cache miss total                     283,464     57.71%      8.59GB
>> 61.05%
>>
>> Client aborted                         8,789      1.79%      4.93GB
>> 35.03%
>> Connect failed                        46,705      9.51%     21.14MB
>> 0.15%
>> Invalid request                      133,732     27.23%     24.28MB
>> 0.17%
>> Unknown error(99)                      1,449      0.29%      0.00KB
>> 0.00%
>> Other errors                               0      0.00%      0.00KB
>> 0.00%
>> Errors total                         190,675     38.82%      4.97GB
>> 35.35%
>>
>> ..............................................................................
>> Total requests                       491,187    100.00%     14.07GB
>> 100.00%
>>
>> as you can see above statistics, *Client-aborted* is taking many
>> *Byte-Percent* and *Invalid-request* is taking many *Count-percent*.
>> do you have any idea about this? are there any config parameters to
>> reduce these error counts?
>>
>> i change some configs like below:
>>
>> ##############################################################################
>> # HTTP connection timeouts (secs). Docs:
>> #
>> https://docs.trafficserver.apache.org/records.config#http-connection-timeouts
>>
>> ##############################################################################
>> #CONFIG proxy.config.http.keep_alive_no_activity_timeout_in INT 115
>> CONFIG proxy.config.http.keep_alive_no_activity_timeout_in INT 3600
>> #CONFIG proxy.config.http.keep_alive_no_activity_timeout_out INT 120
>> CONFIG proxy.config.http.keep_alive_no_activity_timeout_out INT 900
>> CONFIG proxy.config.http.transaction_no_activity_timeout_in INT 30
>> CONFIG proxy.config.http.transaction_no_activity_timeout_out INT 30
>> #CONFIG proxy.config.http.transaction_active_timeout_in INT 900
>> CONFIG proxy.config.http.transaction_active_timeout_in INT 14400
>> #CONFIG proxy.config.http.transaction_active_timeout_out INT 0
>> CONFIG proxy.config.http.transaction_active_timeout_out INT 14400
>> CONFIG proxy.config.http.accept_no_activity_timeout INT 120
>> CONFIG proxy.config.net.default_inactivity_timeout INT 86400
>>
>> but it didn't removed the problem completely.
>>
>>
>> --
>> Best Regards,
>>     R. Arasteh
>>
>>
>


-- 
Best Regards,
    R. Arasteh

Mime
View raw message